臺灣博碩士論文加值系統

本研究應用支持向量機及線性回歸演算法，對台中市某都市污水處理廠出流水水質進行分析並預測，透過蒐集污水處理廠的進流水各項水質數據及操作參數，探討進流水數據對出流水之影響。
本研究對台中市某都市污水處理廠出流水水質建立支持向量機及線性回歸模型。其結果顯示，出流水pH預測值支持向量機之MAPE (Mean absolute percentage error)值最佳為0.68%，線性回歸之MAPE值最佳為0.59%。出流水水溫預測值支持向量機之MAPE值最佳為0.80%，線性回歸MAPE最佳為0.77%。出流水BOD5預測值支持向量機之MAPE值最佳為19.22%，線性回歸之MAPE值最佳為20.11%。出流水COD預測值支持向量機之MAPE值最佳17.56%為，線性回歸之MAPE值最佳為20.60%。出流水DO預測值支持向量機之MAPE值最佳5.20%為，線性回歸之MAPE值最佳為5.45%。出流水SS預測值支持向量機之MAPE值最佳為4.01%，線性回歸之MAPE值最佳為4.92%。出流水導電度預測值支持向量機之MAPE值最佳為1.75%，線性回歸之MAPE值最佳為1.96%。
本研究利用支持向量機及線性回歸模型，預測都市污水廠放流水水質，透過相關係數篩選相關性高的數據進行分析，根據結果顯示支持向量機在預測水溫、pH、DO、SS、導電度皆為高精確；預測COD及BOD5為優良。線性回歸預測水溫pH、DO、SS、導電度皆為高精確；BOD5及COD為合理。兩預測模型皆在預測出流水水質上有一定的效果。

This study applies support vector machine and linear regression algorithms to analyze and predict the effluent water quality of an urban wastewater treatment plant in Taichung City. By collecting various water quality data and operational parameters of the influent water at the treatment plant, the study investigates the impact of influent data on effluent water quality.
The study establishes support vector machine and linear regression models for predicting the effluent water quality of the urban wastewater treatment plant. The results indicate that the best MAPE (Mean Absolute Percentage Error) value for predicting effluent pH using the support vector machine is 0.68%, while the best MAPE value using linear regression is 0.59%. For predicting effluent water temperature, the best MAPE value using the support vector machine is 0.80%, and the best MAPE value using linear regression is 0.77%. The best MAPE value for predicting effluent BOD5 using the support vector machine is 19.22%, while the best MAPE value using linear regression is 20.11%. For predicting effluent COD, the best MAPE value using the support vector machine is 17.56%, and the best MAPE value using linear regression is 20.60%. The best MAPE value for predicting effluent DO using the support vector machine is 5.20%, while the best MAPE value using linear regression is 5.45%. For predicting effluent SS, the best MAPE value using the support vector machine is 4.01%, and the best MAPE value using linear regression is 4.92%. The best MAPE value for predicting effluent conductivity using the support vector machine is 1.75%, while the best MAPE value using linear regression is 1.96%.
This study utilizes support vector machine and linear regression models to predict the effluent water quality of urban sewage treatment plants. By screening highly correlated data through correlation coefficients for analysis, the results demonstrate that the support vector machine achieves high precision in predicting water temperature, pH, DO, SS, and conductivity, with excellent predictions for COD and BOD5. Linear regression also achieves high precision in predicting water temperature, pH, DO, SS, and conductivity, with reasonable predictions for COD and BOD5. Both prediction models demonstrate a certain level of effectiveness in predicting effluent water quality.

目錄 i
表目錄 ii
圖目錄 iii
第一章 緒論 1
第一節 研究背景與動機 1
第二節 研究目的 3
第三節 名詞釋義 3
第二章 文獻回顧 5
第一節 都市污水廠 5
第二節 支持向量機 10
第三節 線性回歸 15
第三章 研究方法 19
第一節 研究架構與流程 19
第二節 支持向量機 24
第三章 線性回歸 29
第四節 模式預測與分析 33
第五節 模式預測效能之評估 34
第六節 R語言 38
第四章 結果與討論 41
第一節 支持向量機與線性回歸模型參數篩選 41
第二節 支持向量機模型預測結果 47
第三節 線性回歸預測結果 69
第五章 結論與建議 91
第一節 結論 91
第二節 建議 93
參考文獻 94

白子易（2001）。下水道系統生化動力模式建立之研究。﹝博士論文。國立中央大學﹞臺灣博碩士論文知識加值系統。https://hdl.handle.net/11296/5j7w46。
李孟諵(2011)。R-Project之初探。SRDA學術調查研究資料庫通訊。36。11–19。https://srda.sinica.edu.tw/filepool/Tbemagzineflatt/51.pdf
林英才、林雪良、李嘉展、彭清勇、聶方珮、鄭明得（2015）。牛樟菇菌最適化固態培養條件與化學成分之探討。台北海洋技術學院學報，6(2)，58-71。https://www.airitilibrary.com/Article/Detail?DocID=20713630-201503-201511230029-201511230029-58-71
林紘原（2005）。廢水處理調勻池效能評估指標建立之研究。台灣環境資源永續發展學會。http://www.tasder.org.tw/meeting/2005/A07-%E5%BB%A2%E6%B0%B4%E8%99%95%E7%90%86%E8%AA%BF%E5%8B%BB%E6%B1%A0%E6%95%88%E8%83%BD%E8%A9%95%E4%BC%B0%E6%8C%87%E6%A8%99%E5%BB%BA%E7%AB%8B%E4%B9%8B%E7%A0%94%E7%A9%B6-%E6%9E%97%E7%B4%98%E5%8E%9F-%E5%85%A8%E6%96%87.pdf
經濟部水利署（2023）。自來水生活用水統計。
https://www.wra.gov.tw/News_Content.aspx?n=2868&s=6999
臺灣環保暨資源再生設備工業同業公會（2022）。水/廢水處理與再生。臺灣環保暨資源再生設備工業同業公會。https://tema.org.tw/tw/news-detail-8-890.html
蔡清讚（1979）。活性污泥系統電腦模擬分析。﹝碩士論文。國立臺灣大學﹞臺灣博碩士論文知識加值系統。 https://hdl.handle.net/11296/atcgch。
藍啟裕（2019）。單變量時間序列預測模型之比較以台中市空氣污染PM2.5指數為例。﹝碩士論文。國立中興大學﹞臺灣博碩士論文知識加值系統。 https://hdl.handle.net/11296/r94uwq。
Ali, A, M., Fazlollah, C., Izharul, H. F., Khan, S. U. K., Mahmood, Y., Nadeem, A., & Sirajuddin, A. (2020). Recent trends in disposal and treatment technologies of emerging- pollutants- A critical review. TrAC Trends in Analytical Chemistry, Volume 122, 2020, 115744, ISSN 0165-9936, https://doi.org/10.1016/j.trac.2019.115744.
Aytaç, A., & Seçkin, K. (2019). The Effect of Kernel Values in Support Vector Machine to Forecasting Performance of Financial Time Series and Cognitive Decision Making. THE JOURNAL OF COGNITIVE SYSTEMS, 4(1), 17–21.
Ben, L., Bolan, L., Fanshuo, L., Junwei, Z., & Peng, W. (2022). Fault mode detection of a hybrid electric vehicle by using support vector machine.Energy Reports, Volume 9, Supplement 8, 2023, Pages 137-148, ISSN 2352-4847, https://doi.org/10.1016/j.egyr.2023.04.328.
Bodaka H, Farhoud N, Hlali E. (2023). Modeling of wastewater treatment plant in Hama city using regression and regression trees. Environ. Health Eng. Manag. 2023; 10 (3) :293-300.
Boser, B. E., Guyon, I. M., & Vapnik, V. N. (1992). A training algorithm for optimal margin classifiers. Proceedings of the fifth annual workshop on Computational learning theory – COLT '92, 1992, 144, ISBN 089791497X, doi:10.1145/130385.130401
Bun, Y., Hiroshi, Y., Jinichi, H., Kazuno, N., Masaru, M., Motoki, S., & Yasuharu, Y.(2020). Machine-learning approach to predict on-road driving ability in healthy older people, Psychiatry and Clinical Neurosciences. Volume 74, Issue 9 , p. 488-495
Cortes, C. & Vapnik, V. (1995). Support-vector networks. Machine Learning. 1995, 20 (3), 273–297. doi:10.1007/BF00994018.
Cristianini, N., & Shawe-Taylor, J. (2000). An Introduction to Support Vector Machines and other kernel-based learning methods. Cambridge University Press. ISBN 0-521-78019-5.
Cortez, P. (2012). Data Mining with Multilayer Perceptrons and Support Vector Machines.. Intelligent Systems Reference Library, vol 24. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-23241-1_2
Gigante, D., Oliveira, P., Fernandes, B., Lopes, F., Novais, P. (2021). Unsupervised Learning Approach for pH Anomaly Detection in Wastewater Treatment Plants. In: Sanjurjo González, H., Pastor López, I., García Bringas, P., Quintián, H., Corchado, E. (eds) Hybrid Artificial Intelligent Systems. HAIS 2021. Lecture Notes in Computer Science(), vol 12886. Springer, Cham. https://doi.org/10.1007/978-3-030-86271-8_49
Glory, V. P., Jagupilla, L. P., Mokkapati, R. K., & Santhosh, C. (2023). Static Hand Gesture Recognition Using Novel Convolutional Neural Network and Support Vector Machine. International Journal of Online and Biomedical Engineering (iJOE), 19(09), pp. 131–141. https://doi.org/10.3991/ijoe.v19i09.39927
Gregório, J.; Gouveia-Caridade, C.; Caridade, P.J.S.B. (2022). Modeling PM2.5 and PM10 Using a Robust Simplified Linear Regression Machine Learning Algorithm. Atmosphere 2022, 13, 1334. https://doi.org/10.3390/atmos13081334
Haeng Yeol Oh, Myeong-Hun Jeong, Seung Bae Jeon, Tae Young Lee, Gun Kim, Minkyo Youm (2021) Sea Water Quality Estimation Using Machine Learning Algorithms. Journal of Coastal Research 1 September 2021; 114 (SI): 424–428: https://doi.org/10.2112/JCR-SI114-086.1
Hoon, S., Hua, W., Lodewijk, B., & Lucia, S. B. (2022). Scaling multi-instance support vector machine to breast cancer detection on the BreaKHis dataset. Bioinformatics, Volume 38, Issue Supplement_1, July 2022, Pages i92–i100, https://doi.org/10.1093/bioinformatics/btac267
Kazi, Z.; Filip, S.; Kazi, L. (2023). Predicting PM2.5, PM10, SO2, NO2, NO and CO Air Pollutant Values with Linear Regression in R Language. Appl. Sci. 2023, 13, 3617. https://doi.org/10.3390/app13063617
Liu, Zj., Wan, Jq., Ma, Yw. et al. (2019). Online prediction of effluent COD in the anaerobic wastewater treatment system based on PCA-LSSVM algorithm. Environ Sci Pollut Res 26, 12828–12841 (2019). https://doi.org/10.1007/s11356-019-04671-8
M. Botros (2006). Evolving complex robotic behaviors using genetic programming. In Genetic Systems Programming, Volume 13 of Studies in Computational Intelligence, eds. by A. Abraham, N. Nedjah, L. de Macedo Mourelle, (Springer, Berlin, 2006), pp. 173–191
Md. Saikat Islam Khan, Nazrul Islam, Jia Uddin, Sifatul Islam, Mostofa Kamal Nasir(2022). Water quality prediction and classification based on principal component regression and gradient boosting classifier approach, Journal of King Saud University - Computer and Information Sciences, Volume 34, Issue 8, Part A,2022,Pages 4773-4781,ISSN 1319-1578, https://doi.org/10.1016/j.jksuci.2021.06.003.
Okeke OP, et al. (2022) Performance analysis and control of wastewater treatment plant using Adaptive Neuro-Fuzzy Inference System (ANFIS) and Multi-Linear Regression (MLR)techniques. GSC Adv Eng Technol 2022:1-16.
Onkar Ghadge, & Vikas Ghute (2023). GROUP RUNS AND MODIFIED GROUP RUNS CONTROL CHARTS FOR MONITORING LINEAR REGRESSION PROFILES. Reliability: Theory & Applications, 18 (4 (76)), 513-524. doi: 10.24412/1932-2321-2023-476-513-524
Pai, T.Y., Chuang, S.H., Wan, T.J. et al. (2008) Comparisons of grey and neural network prediction of industrial park wastewater effluent using influent quality and online monitoring parameters. Environ Monit Assess 146, 51–66. https://doi.org/10.1007/s10661-007-0059-7
Pai T.Y., Lo H.M., Wan T.J., Chen L., Hung P.S., Lo H.H., Lai W.J. and Lee H.Y. (2015). Predicting air pollutant emissions from a medical incinerator using grey model and neural network. Applied Mathematical Modelling, 39 (5-6), 1513-1525. 10.1016/j.apm.2014.09.017.
Pai T. Y., Sung P.J., Lin C.Y., Leu H.G., Shieh Y.R., Chang S.C., Chen S.W. and Jou J.J. (2010). Predicting hourly ozone concentration in Dali area of Taichung County based on multiple linear regression method. International Journal of Applied Science and Engineering, 7(2), 127-132.
Pai T.Y., Wu R.S., Chen C.H., Chen L., Lin C.Y., Lee H.Y., Shih L.H., Jiang Y.Z. and Shen C.Y. (2014). Predicting hardness of four groundwater monitoring stations in Kaohsiung City of Taiwan using seven types of GM (1, 1) model. Advanced Materials Research, 905, 314-317.
Qichao Zhao , Xuxin Dong, Guohong Li , Yongtao Jin , Xiufeng Yang , and Yandan Qu. (2022). Classification and Regression Tree Models for Remote Recognition of Black and Odorous Water Bodies Based on Sensor Networks. Scientific Programming, 2022, 1–12. https://doi.org/https://doi.org/10.1155/2022/7390098.
Ribeiro, D., Sanfins, A., Belo, O. (2013). Wastewater Treatment Plant Performance Prediction with Support Vector Machines. In: Perner, P. (eds) Advances in Data Mining. Applications and Theoretical Aspects. ICDM 2013. Lecture Notes in Computer Science, vol 7987. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-39736-3_8
Seung PL, Sang YM, Jin SK, Jong U, Man SK (2014) A study on the influence of a sewage treatment plant’s operational parameters using the multiple regression analysis model. Environ Eng Res 19:31–36
Smita Rath, Alakananda Tripathy, Alok Ranjan Tripathy, Prediction of new active cases of coronavirus disease (COVID-19) pandemic using multiple linear regression model, Diabetes & Metabolic Syndrome: Clinical Research & Reviews, Volume 14, Issue 5, 2020, Pages 1467-1474, ISSN 1871-4021, https://doi.org/10.1016/j.dsx.2020.07.045.
T.Y. Pai, P.Y. Yang, S.C. Wang, M.H. Lo, C.F. Chiang, J.L. Kuo, H.H. Chu, H.C. Su, L.F. Yu, H.C. Hu, Y.H. Chang, (2011). Predicting effluent from the wastewater treatment plant of industrial park based on fuzzy network and influent quality, Applied Mathematical Modelling, Volume 35, Issue 8, 2011, Pages 3674-3684, ISSN 0307-904X, https://doi.org/10.1016/j.apm.2011.01.019.
T.Y. Pai, T.J. Wan, S.T. Hsu, T.C. Chang, Y.P. Tsai, C.Y. Lin, H.C. Su, L.F. Yu, (2009). Using fuzzy inference system to improve neural network for predicting hospital wastewater treatment plant effluent, Computers & Chemical Engineering, Volume 33, Issue 7, 2009, Pages 1272-1278, ISSN 0098-1354, https://doi.org/10.1016/j.compchemeng.2009.02.004.
T.Y. Pai, Y.P. Tsai, H.M. Lo, C.H. Tsai, C.Y. Lin, (2007). Grey and neural network prediction of suspended solids and chemical oxygen demand in hospital wastewater treatment plant effluent, Computers & Chemical Engineering, Volume 31, Issue 10, 2007, Pages 1272-1281, ISSN 0098-1354, https://doi.org/10.1016/j.compchemeng.2006.10.012.
Vignesh, p. p., Jiang, j. h., & Kishore, p. (2023). Predicting PM2.5 Concentrations Across USA Using Machine Learning, Earth and Space. ScienceVolume 10, Issue 10, https://doi.org/10.1029/2023EA002911
Wang,F.,Chen,W.,Fakieh,B. & Ali,B.(2022).Stock price analysis based on the research of multiple linear regression macroeconomic variables. Applied Mathematics and Nonlinear Sciences,7(1) 267-274. https://doi.org/10.2478/amns.2021.2.00097
W.C. Leong, R.O. Kelani, Z. Ahmad, (2020). Prediction of air pollution index (API) using support vector machine (SVM), Journal of Environmental Chemical Engineering, Volume 8, Issue 3, 2020, 103208, ISSN 2213-3437, https://doi.org/10.1016/j.jece.2019.103208
Zhang, Q. (2021). Housing Price Prediction Based on Multiple Linear Regression. Scientific Programming 2021(1), 1–9, https://doi.org/https://doi.org/10.1155/2021/7678931
Yan Zheng Liu, Zhiyuan Chen; (2023). Prediction of biochemical oxygen demand with genetic algorithm-based support vector regression. Water Quality Research Journal 1 May 2023; 58 (2): 87–98. doi: https://doi.org/10.2166/wqrj.2023.015
Z. Muhammad, N. A. J. Jailani, N. A. M. Leh and S. A. Hamid. (2022). Classification of Drinking Water Quality using Support Vector Machine (SVM) Algorithm. IEEE 12th International Conference on Control System, Computing and Engineering (ICCSCE), Penang, Malaysia, 2022, 75-80, 10.1109/ICCSCE54767.2022.9935657.